Planet of the Bugs: Evolution and the Rise of Insects

by Scott Richard Shaw

  Although flight allowed the mayflies to rise in the air, easily find partners, and breed in relative safety, wings also served another valuable function: they helped mayflies to disperse. Nymphs that develop in streams tend to get washed downstream as they feed and develop, and by the time they emerge into adulthood, they are some distance from the egg-laying sites. With their wings, early mayflies could have moved easily upstream, laid eggs, and colonized inland lakes and ponds, which had fewer fish than the deeper waters.

  Coal Country Tour

  When those first mayflies took to the air some 320 million years ago, they glittered and danced over vast lowland marshes, swamps, and tropical wet forests covered with giant horsetails, club mosses, ferns, now-extinct seed fern trees, cordaites plants, and ancient conifers.4 These forests were unlike any that exist today, yet we know them well because they left behind abundant fossil plants. The accumulations of these plant materials—the most prolific in the history of life—became most of our natural gas, oil, petroleum, and especially our coal, which are remnants of trees that grew to heights of thirty feet and more in dense wet forest stands. During storms these large trees fell into the moist Carboniferous soils and swamps. Their remains were flooded and buried under sediments, piling up to form deep layers of carbon-rich organic debris. Over time the pressure of geological activity transformed the layers into coal. My university generates electricity from a coal-fired facility, so as I write these words I’m using power produced from some of the last vestiges of Carboniferous plant life. When you drive your car to the supermarket, you might be using some as well.

  Why does most of our coal and much of our petroleum date from the Carboniferous period? The orthodox view is simply that the Carboniferous swamps provided optimal conditions for fossil fuel formation. After that time, the world became drier, and conditions were not as favorable. But is that all there is to it? Forests didn’t go away after the Carboniferous. If anything, there were even more trees, growing ever more widely at higher elevations, and growing to greater heights over the Mesozoic and Cenozoic. After the Carboniferous, the continents may have been drier inland, but they still had lots of streams, rivers, and coastal wetlands. Trees from dry highlands could easily have washed into rivers and accumulated in lowland marshes. Something other than a change in the weather must have occurred.

  Let’s consider another question: what happens to a tree when it dies in a modern forest? Birds, such as woodpeckers, chickadees, and nuthatches, excavate cavities in it. When these birds move out, lots of small mammals move in and excavate even larger spaces. Other smaller animals cut tunnels in the dead wood: bees and wasps, for instance, do so not to seek food but purely to build nesting sites. The tree’s bark is attacked and colonized by various insects, such as bark beetles, flat-headed wood-borers, and gnawing bark lice. Their tunneling activity loosens the bark and allows fungi to spread under that surface. Fungal growth speeds the tree’s decomposition and provides an even more nutritious food source for insects than the wood itself. Larger species of wood-boring beetles lay their eggs on the tree, and their grublike larvae tunnel deep into the heartwood. Meanwhile, down below, bacterial and fungal growth decompose the roots, which small soil arthropods chew on. Eventually the roots are weakened and the tree falls in a storm. This totally changes the local environmental conditions, since much of the tree is now spread across moist soil. It is even more exposed to fungi and small soil arthropods. In tropical areas, the termites move in and tear the wood to pieces. They, along with the wood roaches, are among the few animals capable of digesting wood because of the symbiotic microorganisms living in their guts. In temperate areas as well as the tropics, the carpenter ants will likely move in as well. They do not eat the wood; they simply tunnel and live inside it, but eventually they reduce a large tree to small fragments and wood dust. The pieces are mixed with soil and further reduced by soil fungi and micro-arthropods. In modern forests, there isn’t much left that might survive and fossilize into coal or petroleum. Most of the tree is recycled back into the forest’s living systems.

  In the Early Carboniferous, most of today’s macroscopic and microscopic consumers of dead wood had not yet evolved.5 There were no birds, mammals, bees, wasps, bark beetles, wood-boring beetles, bark lice, termites, or ants. Moreover, during the Devonian and Carboniferous times, plants became tall by producing cellulose and lignin, which are very difficult for animals to digest. None of the earliest insects were able to digest raw wood as well. The giant Carboniferous horsetails, like modern horsetails, toughened their vascular tissues with large amounts of silica, making them virtually indigestible. So the Late Devonian and Carboniferous really were special for their excess production of plant materials, not only because the moist climate and high levels of atmospheric carbon dioxide favored plant growth, but also because the plants were able to produce more biomass than the herbivores could consume, for millions of years. The first important insect wood consumers—the wood roaches—did not appear until the Late Carboniferous. They were followed by the appearance of bark lice and the diversification of wood-boring beetles in the Permian.6 Over time, increasingly more complex communities of wood consumers evolved, and the global bulk production of plant materials of the Carboniferous has never been repeated.7

  Maiden Flight

  The most notable of the new kinds of vertebrate predators that appeared in the Early Carboniferous are the so-called keyhole amphibians, named after their peculiar keyhole-shaped eye sockets. They were the first four-legged vertebrates to evolve ears. What sounds were these amphibians listening to? I’ve read that they might have developed sound-producing capabilities and used songs to attract mates or mark territories. That may be the case, but mating is just one aspect of mature vertebrate animal behavior. Ears would have been very handy for more mundane aspects of daily life, like finding arthropod food. The large amphibians may have eaten fish and other amphibians, but the small ones would certainly have eaten a wide variety of arthropods and insects. With ears they could have heard the rustling movements of arthropods moving through the leaf litter, the chewing sounds of millipedes and insects crunching on their meals, or the fluttering motions of newly emerged insects preparing to fly. It’s probably no coincidence that vertebrates developed ears at a time when arthropods were starting to make a lot of buzzing and fluttering noises in the forest. The next time you listen to your favorite tunes, perhaps you should take a moment to appreciate the fact that our Carboniferous ancestors liked to eat bugs.

  The forests of the Carboniferous years produced more than just our ears and coal supplies. By the Pennsylvanian subperiod (the latter half of the Carboniferous, about 320 million years ago), they housed the very first flying insects. The first step in understanding the origins of insect flight is to consider the ancestral organisms from which flying things evolved. During the Carboniferous there appeared a simple order of insects called Zygentoma, known to us as silverfish and firebrats.8 Zygentomans are considered the nearest relatives of flying insects, and as their common names suggest, some species have silvery scales, like fish, while others have an affinity for warm places (some are pests in bakeries and live near ovens). Like modern insects, they have a head, a thorax with six legs, and a long abdominal region, and similar to bristletails, the Zygentoma have three long filamentous appendages trailing from the end of their body. They do not have wings or even wing buds, but the upper parts of each thoracic segment are flattened and extend slightly to the sides. Ancient zygentomans probably climbed up on emergent vegetation to molt, and it’s been suggested that they could probably jump from their lofty perches, just like modern silverfish.

  Zygentoma share some unique features with flying insects, including a jaw that’s hinged on two condyles, two cerci, and an infolding of the thorax’s side wall—the pleuron—which produces a pleural suture. This is a crucial trait leading to wing development because it both produces a structurally stronger thoracic area and provides more internal room for muscle attachment.
But the similarity to flying insects ends there; zygentoman nymphs resemble tiny versions of an adult, and neither ever shows any trace of wing development. The Carboniferous silverfish, however, did have slightly expanded lobes along the upper sides of their thoracic segments, which some insect paleontologists have interpreted as protowings. They certainly did not have wings, but they may have been capable of brief gliding flight after jumping or falling from perches on high vegetation.9

  Some fossilized insects from the Carboniferous have flat, platelike outgrowths along the upper sides of the thorax called paranotal lobes. These lobes are larger than those of Zygentoma, but the presumption is that they are modifications of the same structures. In some cases, paranotal lobes are present on the first segment of the thorax, while fully formed wings are present on the middle and last segments. The implication is that wings developed from such lobes, which later disappeared for functional reasons. No modern insect has wings, winglets, or lobes on the first thoracic segment. Instead, their wings are located on the middle or last ones.10 Winglets disappeared from the front end of the thorax probably because an insect’s center of gravity is located more toward its rear, and its first segment is too small to allow extensive musculature.

  Because wings appeared at the same time that forests became widespread, some entomologists assume that insects evolved flight by climbing tall plants, jumping off, and gliding with their flat plates—an idea known as the paranotal lobe hypothesis. But why would soil insects, which were perfectly safe and happy in the leaf litter, bother to climb way up on top of plants? Entomologists usually suggest that they may have been feeding on the developing spores, seeds, and tender photosynthetic tissues. This idea troubles me for a couple of reasons. First, modern silverfish don’t feed in that way. Second, spores and seeds would fall to the ground anyway, so why bother climbing? I can think of a number of better reasons why insects might have climbed plants, even if they didn’t feed there.

  Insects are cold-blooded. If nighttime temperatures are chilly, their bodies cool down, and they don’t get moving again until they warm up. Modern insects often solve this problem by climbing plants, perching in the sunshine, and using their wings as solar panels. The larger structural veins of the wings are hollow, so blood flows into them, allowing heat to transfer back into the body. Even small protowings would have had the potential to transfer valuable heat, possibly before the panels could be used for flight. Any insect needs to warm up first before it can walk, gather food, search for mates, or avoid predators, and whoever starts moving first holds the advantage. This solar panel hypothesis for insect flight is particularly compelling when you consider the prevailing climate of the Carboniferous. We know that the Devonian ended with a global cooling period and that glaciers formed over much of the southern continents during the Carboniferous. Tropical regions remained moderate and free of ice, but the average climate would have been much cooler than during the Devonian. There needs to be some explanation of why soil-dwelling insects would leave the comfort and safety of the soil layer, and the quest for sunlight and heat might provide a plausible solution.

  Or maybe they were looking for mates. Some male insects wait for females at feeding sites (the dinner and dancing, disco-bar strategy), while others wait for young virgin females near emergence sites (the Lolita strategy). When these tactics don’t work, lots of males go to a prominent landmark (the hilltop lover’s lane strategy). Many modern insects move to the highest tree, rock, or hill in the area, so it’s possible that Carboniferous insects met mates by climbing to the tops of the tallest plants.

  Once on those plants, small protowings possibly played a role in courtship and mating. We know that some modern insects produce wing vibrations and use these sounds as courtship signals and that others use their wings to actively disperse chemical scents called pheromones, which signal and attract mates; even a small protowing might have been used in these ways. Moreover, fossil impressions indicate that some of the Carboniferous insects had distinctive pigmentation patterns on their wings. In some instances, these patterns included dorsally visible spots or disruptive banding that may have served to warn off predators. We can’t be sure of these patterns’ exact function, but it’s logical to assume that ancient winged insects, like modern ones, also used them to attract potential partners from a distance.

  For whatever reasons ancient insects might have climbed plants, once they were up there, they needed to get down again, and what simpler way than to jump? Some tropical frogs climb trees in the evening to feed in the forest canopy then return to the ground at dawn by jumping and gliding. If frogs can safely glide from treetops, then why not insects, which are so small that most could fall some distance without harm? Some scientists have suggested that because the Carboniferous atmosphere was dense with oxygen and carbon dioxide, insects could easily learn to fly. Others have suggested that even the bristletails’ tiny thoracic lobes might have provided them with simple gliding flight capability.

  On the other hand, some entomologists have suggested that insects didn’t need to climb plants to learn how to fly: instead, wings might have evolved from gills and played a role in breathing. This is an old idea that enjoyed popularity more than a century ago. In fact, entomologists developed the paranotal lobe hypothesis to challenge it and explain how wings might have evolved without involving respiration, since modern insect wings do not have any respiratory function. But over recent decades the gill hypothesis for wings has gained some new supporters and maintained opposition. Let’s consider its pros and cons. The large, hollow veins of modern insect wings do contain tracheal tubes, so gas as well as blood flows through them. The mayflies, as well as fossil insect wings from the Carboniferous, show us that the most ancient flying insects had enormous numbers of wing veins; presumably they had far more wing tracheae than modern insects. We also know that the two most primitive surviving groups of old-winged insects, the mayflies and the dragonflies, both have aquatic immature stages that breathe with gills, which work by allowing gasses to transfer across a thin moist cuticle. Some soil arthropods, like the springtails, can respire through their cuticle because it too is very thin; perhaps ancient insects might also have breathed across the thin cuticle of their winglets. Because the Carboniferous coal swamps were very moist and the air’s oxygen level was exceptionally high, it does seem possible that the ancient insects’ protowings served as gills before evolving into wings.

  In 1994, Penn State’s Jim Marden, along with his colleagues, suggested a novel idea—the surface-skimming hypothesis—based on the observation that some adult stoneflies (Plecoptera) can use their wings to sail across water. They proposed that ancient aquatic insects might have first used protowings or gills for surface skimming, and later evolved the capacity for powered flight by fluttering up from the water’s surface. Even so, the gill hypothesis has some serious problems. Most notably, although modern insects may have wing tracheae, none are known to actually breathe with them. Instead, when insects molt to the adult stage, air flows into the tracheae and helps pump up the wings to full size. There are other difficulties as well. While surface skimming has been observed among the stoneflies, it has not been observed among the aquatic mayflies and dragonflies, whose gills are on their abdomen, not on their thorax. Silverfish, the most likely ancestors from which winged insects evolved, dwell in the leaf litter and soil. Also, one of the most successful groups of Carboniferous flying insects, the now-extinct Paleodictyoptera, had terrestrial nymphs.11

  Personally, I’m content with the paranotal lobe hypothesis, but the controversy may never be resolved to everyone’s satisfaction. A gap in the fossil record between the early ground-dwelling hexapods in the Late Devonian and the flying insects of the Late Carboniferous renders wing origins all the more mysterious. On the one hand, we need to consider the possibility that Carboniferous conditions enabled early insect wings to respire, but that they lost this ability as insects developed more modern wings. On the other hand, it is apparent that inse
ct wings might easily have developed for several other good, entirely independent reasons: for gliding, heat transfer, or mating display.

  The various hypotheses for wing origin shouldn’t distract us from the bigger picture: wings evolved by the end of the Carboniferous’ Mississippian subperiod (about 327 million years ago), and once they appeared, they became common fairly suddenly in geological terms. Over the course of several million years, flying insects burst into the air with a diversity that exceeded the Cambrian explosion. They have dominated terrestrial ecosystems ever since, so wings were one of the great insect evolutionary innovations that promoted the insects’ disproportionate success in the living world. Keep in mind that they had total command of the air for more than a hundred million years, at least.12 There were no birds yet, and the best any vertebrate predator could do would be to snap at them from the ground. Occupation of the airways provided the insects with brilliant new opportunities for dispersal, escape, courtship, and exploiting feeding sites. Initially, they didn’t have to be very good at flying. For the first plant-feeding insects nibbling on spores at the tops of tall plants, just a short fluttering flight would allow them to easily move from plant to plant. Compared to a crawling insect, which would need to walk all the way back to the ground, find another stem, and climb back up again, a flying insect could save an enormous amount of time and energy. And it would run into fewer predators in the process. Even a gentle flight would allow one to occupy the air and perform a safe mating dance, or move to a new habitat when the swamps dried up or wildfires moved through the area.

  Classic Fashions Never Go Out of Style

  The most ancient wing style—a flat panel of skeletal material set into a membranous area at the top side of the thorax—was very simple but highly efficient, and some modern insects such as mayflies and dragonflies still sport it. The middle of the wing balances on a fulcrum formed from the pleuron. The wing itself acts as a lever, snapping up or down, because it is stable only in the most raised and the most lowered positions. Its upstroke is accomplished indirectly by muscles that connect the thorax’s top and bottom walls but do not attach to the wing. When these dorsoventral flight muscles contract, the thorax is flattened and distorted, causing the wing to snap into an upright position. In the paleopteran insects, the downstroke is accomplished by muscles connected to small plates below the wing, just near the front and back. These are sometimes called direct flight muscles, because they connect directly to the wing. They not only power a downstroke, but by differential contractions they can allow the wing to be tilted at different angles during flight, allowing directed navigation.